Sunday, August 22, 2021

Robo-Advisors versus Financial Advisors


I was thinking about the panel I had with The Artist Formerly Known as Money Maverick or Luke Ho, and he mentioned an interesting tidbit about how much concern robos have been causing to FAs. I was also able to read an entertaining tirade from an angry financial advisor who wrote a nasty FB message against the Syfe product offering ( If I were Syfe legal counsel, I would act against that clown ).  

These two events have caused me to spend the past few days thinking about whether investment trainers can work with robo-advisors closer in the future.  

As of now, I'm not convinced that we should do so. 

Robo are still quite mum about their algorithms and in a place like Singapore, the biggest strength of using a robo to do tax-loss harvesting cannot be realised in our tax regime. Furthermore, my guess is that robos are running on VC funding right now, and would either have to consolidate or raise fees in the future.  

The only reason I like robos is that they make FAs nervous and can potentially wreak havoc with middlemen commissions should they become more mainstream here. 

Here are some random points from my thinking aloud:

  • It is theoretically possible for DIY investment trainers to work with robo-advisors. Not all investors have the time to build a portfolio of their own, and the risk of cannibalization is very small. 
  • DIY investors still make plenty of qualitative decisions that robo-advisors will never be able to replicate. This is important because qualitative differences may drive future alpha generation for either side. Robos will not be able to win over the best retail investors anyway.
  • DIY investment trainers with any backbone would seek more information on the way the robos function. For me, I need to understand it to the point until I can code the program myself. Otherwise, it would be hard to entertain questions from students on how robos work. I got one data point on a S-REIT based robo and for the life of me, I can't even explain why it underperformed so much compared to a REIT portfolio built by my students, almost all beginners. 
  • It would also be useful to know how much will a robo need to charge existing clientele to break even. Surviving on VC burn rate is not sustainable. 
  • My ideal scenario is multiple robos will need to provide algorithmic transparency so that a trainer can explain the differences between each product offering and can show students how to pick a robo without conflict of interest. 
  • The unpleasant alternative is for one robo to sponsor the cost of running a training program, but I won't do it even if they paid me and I won't recommend anyone attend a course like this. You pay me with more transparency and not cash. 

As it stands, anything that makes financial advisors nervous is a good opportunity for the training industry because we know that there is still quite a lot of fat to trim and middle-class Singaporeans cannot continue to always sponsor a caste of commissioned salesmen and pay for their Audis when they also have to sponsor expensive tuition agencies as well.  

If robo-advisors band together to build an association to standardise the rules on transparency, I am open-minded to an alliance if it means more Singaporeans are educated on personal finance. 

Because even America has to get out of Afghanistan after 20 years.


  1. Funny enough I was just reading about the fallibility of over relying on algorithms in investing e.g. using mean variance optimisation aka efficient frontier to select stocks or asset classes. Firstly, even 20 years of data is considered too short & likely results in recency bias. Secondly, all are hindsight-biased & even small variation in inputs result in large variation in outputs, so you're trusting what has happened in the recent past (10-30 years) will be the same in the future. Basically the conclusion is that using algos can work, until it suddenly doesn't --- the main takeaway is not to be shocked but to have a plan when everything goes to hell, becoz 100% it will happen.

    Too much opposing vested interests for robos & FAs to cooperate much. At most FAs are looking more for platforms that can efficiently implement their advice for clients. For robos, it's only those with no "secret sauce" and just using plain old fixed asset allocations like 60:40 or 30:70 etc and relying on common ETFs that charge 0.05%, that will be open to working with FAs.

  2. Where can I see the tirade by the FA? Just curious on what he/she wrote

    1. Sadly, I did not have a recorded copy of it. Maybe join the BIGS FB group and do a search. Investment Moats may have shared a copy.

    2. Never mind then. Do you remember the gist of what the FA wrote?

  3. It was a long rambling tirade targetted at Syfe I think. Sounds very anti-Robo, but entertaining because his points were not written clearly.